[Foundation-l] Moving towards a more usable MediaWiki

Gerard Meijssen gerard.meijssen at gmail.com
Tue Dec 2 13:05:50 UTC 2008


Hoi,
It is hardly relevant what the demographics of the test group is. 100%
failure in Tanzania, well educated people in Austria failing to get to grips
with Wikipedia, we get people informing us about our perceived security
problem. The reason why it is hardly relevant is because the same studies
show that the changes implemented made a measurable difference. There is no
point finding fault at this. Even when a different demographic would be less
disastrously bad, there is a solution that is known to improve the odds of
finding people collaborating on a MediaWiki installation.

When you look at the bottom 80% of our projects, there are few people to
learn things from. The notion that you best learn it on a Wikipedia negates
the fact that I find 80% of our Wikipedia projects struggling. Now consider
when our software is more usable, it becomes even easier to learn what there
is to learn. There will be fewer reasons why people do not contribute
because with improved usability, road blocks are removed.

People do not want to read documentation. They do not want to learn by
example. It should be obvious. Now, I would prefer for the WMF to announce
that they are going to develop WYSIWYG, but perfection is the enemy of the
good. We can implement CreatePage now and WYSIWYG is likely to take more
then a year to develop.  So I advocate to do what can be done now.

When you say that improvements in usability is not the only solution, you
are completely correct. Localisation is another factor that plays an
important role. Lack of infrastructure is another. However, usability is
something that *we can* tackle, the best news is that it will help all our
projects. Localisation is worked on with considerable success at Betawiki.
Having said this, Betawiki needs more help. There is a digital divide but
not all Africans or Asians suffers from it. We have to be ready for the
people who can contribute successfully to any MediaWiki project. We have to,
because this is how we can make a difference for all the projects that we
are currently failing.

If you find it unacceptable that the Swahili Wikipedia is struggling, then
just do not accept it, and let usdo what can be done .
Thanks,
      GerardM

2008/12/2 Marco Chiesa <chiesa.marco at gmail.com>

> On Tue, Dec 2, 2008 at 12:51 PM, Gerard Meijssen
> <gerard.meijssen at gmail.com>wrote:
>
> > Hoi,
> > Over the last weeks I have been rather active in promoting improved
> > usability for the MediaWiki software. What really got me going was
> learning
> > from a Wikimania presentation that a UNICEF usability study done in
> > Tanzania
> > showed that 100% of the test subjects were unable to create a new
> article.
> > UNICEF has created extensions to improve on this, extensions that make a
> > difference. The fact that our usability is poor does not only hurt what
> > some
> > call "minority languages". A professor in Austria I know, a veteran user
> of
> > software, was also hard pressed to collaborate on a wiki.
>
>
> Do you know the demographics of the test subjects, and what exactly is
> intended with creating a new article? Is it just creating a new page or
> creating a page with correct formatting, categories, etc.?
>
> >
> > Our usability hurts all our projects. It hurts our smaller projects
> because
> > they do not have enough content and contributors. It hurts our big
> projects
> > because it excludes large demographies from contributing leading to bias
> > and
> > hurting the NPOV of many articles. When we want to reach out, there is no
> > easier way then by making our software usable.
>
>
> The bigger projects started small, the pioneers helped the newbies
> explaining things, new people arrived, and the projects grew. The smaller
> projects don't thrive for a number of reasons, such as too few people
> interested, competition with a larger project (this happens with some
> regional languages, where often "native" speakers are equally native in
> another bigger language), digital divide,.. Of course this can be a vicious
> circle (being small does not motivate people to join, a pioneer can only do
> so much but if there are no followers it gets hard)
>
> >
> > What I propose is that those projects who are interested in improving
> their
> > usability ask the WMF to work with them on this. Given that usable
> software
> > should be understood, and given that this is somewhat experimental in
> > nature
> > as well, it makes sense that project should localise the extensions first
> > before they qualify for an implementation. NB the CreatePage extension
> has
> > only eight messages.
>
>
> I am convinced that the best way to learn to use MediaWiki is editing
> Wikipedia, possibly a large one. Reading the help pages, asking the more
> expert users, and so on. This helps forming the pioneers, the teachers. The
> the teachers go to the small projects, and teach the newbies there. Looking
> at how the source code of a page really is is much more helpful than
> reading
> software documentation, at least for people that don't come from the
> software world
>
> >
> >
> > What I propose is to have many and frequent updates. We should learn from
> > our experience and consequently move forward carefully but deliberately.
> It
> > is not acceptable that so many of our projects are failing. The UNICEF
> > studies explain why this is, the studies show how to improve on this. We
> > just have to apply the lessons learned. We just have to show that we can
> > apply the lessons learnt.
> >
>
> Maybe some project just fail because they're not that useful, that's
> Darwinian selection. I agree that some project could make a big difference
> (a language like swahili, to say one), and it's definitely unacceptable
> that
> they struggle. Usability of the software is important, but I still don't
> believe it will make a difference alone.
>
> Cruccone
> _______________________________________________
> foundation-l mailing list
> foundation-l at lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>


More information about the foundation-l mailing list